DocumentCode :
3141835
Title :
Text feature selection based on improved adaptive GA
Author :
Su, Shuangquan ; Li, Lei ; Zhao, Qing
Author_Institution :
Comput. Coll., Beijing Univ. of Posts & Telecommun., Beijing, China
fYear :
2011
fDate :
27-29 Nov. 2011
Firstpage :
169
Lastpage :
172
Abstract :
Text feature selection is crucial to improve the speed and accuracy of text categorization, and then improves the effect of complex network information filtering. Considering feature selection as an optimization problem, an improved genetic algorithm (GA) is proposed to fulfill text feature selection in this paper. Traditional adaptive genetic algorithm has poor adaptation. Compared with the traditional adaptive genetic algorithm, the fitness function formula, crossover probability formula and mutation probability formula are improved. We illustrated the validity of the improved genetic algorithm through experiments.
Keywords :
genetic algorithms; information filtering; probability; text analysis; adaptive GA; adaptive genetic algorithm; crossover probability formula; fitness function formula; information filtering; mutation probability formula; optimization problem; text categorization; text feature selection; Accuracy; Telecommunications; Adaptive; Genetic Algorithm; Text Feature Selection;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Natural Language Processing andKnowledge Engineering (NLP-KE), 2011 7th International Conference on
Conference_Location :
Tokushima
Print_ISBN :
978-1-61284-729-0
Type :
conf
DOI :
10.1109/NLPKE.2011.6138188
Filename :
6138188
Link To Document :
بازگشت